Abstract: A content management system generally provides the ability to “weight” various content items by assigning weights to various content, ultimately weighting or skewing a knowledge base towards particular or weighted content during the contextualization process. Weighting nodes and/or weighting a content discovery algorithm may result in content items and concepts represented by the content items being considered more heavily by algorithms creating content groupings. An example method includes generating a knowledge base that includes a plurality of nodes representing a respective plurality of content items and updating the knowledge base based on a received user feedback, where the user feedback corresponds to a weighting for one or more content items. Content groupings are generated based on the updated knowledge base, where the content groupings include subsets of the plurality of content items and the subsets include conceptually similar content items.

Abstract: Disclosed reinforcement learning from human feedback techniques provide reward and loss functions for human raters to improve the quality of the feedback data. A certain number of credits are provided to a human rater. The human rater allocates credits to output options that reflect a response or responses to a prompt. The allocations of the credits represent a confidence-weight label for the output options and collectively represent user feedback on the output options. The human rater is rewarded points when the user feedback decreases uncertainty of an AI model. Conversely, the human rater can lose points when the user feedback does not decrease uncertainty of the AI model. A reward model is generated and/or trained based on the user feedback. The AI model is trained based on the reward model.

Abstract: An example method disclosed herein includes receiving a request for content items relevant to a concept within a first knowledge space, where the first knowledge space includes a plurality of concept groupings of a plurality of nodes representing a plurality of content items. The method further includes identifying a plurality of content items in a second knowledge space relevant to the concept and receiving a selection of one or more of the content items for addition to the first knowledge space. The method further includes adding the selected content items to the first knowledge space.

Abstract: A method of identifying one or more concepts in a multimedia file includes separating text derived from the multimedia file into sub-portions, extracting features from the text of the sub-portions, and identifying concept clusters for the sub-portions based on the extracted features. The method further includes associating each of the sub-portions with the one or more concepts presented in the sub-portions of text based on the identified concept clusters and presenting, via a user interface, one or more portions of the multimedia file, where the portions of the multimedia file are generated based on the one or more concepts presented in each of the sub-portions of text of the multimedia file.

Abstract: There is provided a learning management system comprising: a display; and a processing element in electronic communication with the display, the processing element configured to perform the following operations: generate a plurality of nodes, wherein the nodes correspond to an item of content; implement a testing module that selects a first node of the plurality of nodes to display on the display, wherein the first node outputs to the display one or more questions to a user and prompts the user to select an answer; and implement a confidence metric that prompts the user to select a value corresponding with the user's confidence that the selected answer is correct, wherein the selected answer and the confidence metric value for the first node are used to determine a second node from a plurality of nodes to display to the user next.

Abstract: A learning system capable of adaptively presenting various learning materials of various concepts to a user may include a concept tracker to analyze the learning materials and track one or more concepts from each of content items in the learning materials. A content item may be a video, and the system may analyze the transcripts of the video to determine one or more concepts about the video. The learning system may recommend a concept for the user to learn or improve based on the user's understanding and knowledge level around relevant concepts. Further, the system may present learning materials about the recommended concept to the user by indexing video segments by the time stamps associated with the particular concept and replay the video segments from the respective starting position and the ending position without replaying the entire video contents. Various clustering methods may be used to track the concepts.

Abstract: An educational system may include a contextualizer that accesses and processes content items and build a knowledge base including educational content and relationships between content items. The system may include a user assessor unit that determines user assessment variables based on user responses to previous educational content items and generate a user model based on the user assessment variables. The system may include a recommender that navigates the knowledge base and generates recommendations of new educational content based on the user model. The recommendation can assess the user's proficiency on a given concept based on the user model and maximize the user's probability of success around the concept of which the user shows proficiency, or improve the user's proficiency around the concept the user shows weakness. In some examples, the system may acquire an expert's knowledge about certain concepts to refine the user model, increasing the accuracy of the recommender.

Abstract: A method for delivering educational content to a user and detecting cheating in a remote assessment environment. The method includes receiving user assessment variables corresponding to a user performance on assessment questions presented by the educational content; generating a user profile based on the user assessment variables; recommending a new educational content item based on the user profile; presenting the new educational content item to the user via a display; receiving a user response to the new educational content; detecting anomalies in the user response by comparing the user profile with one or more of active user assessment variables or passive user assessment variables; and generating a notification of cheating based on the detected anomalies.